Getting the end of transcripts only

[gwendall]

I am using the listen.live method to get transcripts but am just getting the last words of my speech as transcripts with speech_final: true. It often just returns the last 2-3 words. What did I do wrong? How can I fix it?

Here is my deepgram config:

const deepgramClient = createClient(DEEPGRAM_SECRET_KEY);
const connection = deepgramClient.listen.live({
  model: "nova-2-general",
  encoding: 'linear16',
  language: 'en',
  smart_format: true,
  interim_results: true,
  vad_events: true,
  punctuate: true,
  endpointing: 800,
  utterance_end_ms: 2000,
  minimum_utterance_length_ms: 200
});

And here is how i consider a transcript valid to be handled:

connection?.on(LiveTranscriptionEvents.Transcript, (data: LiveTranscriptionEvent) => {
  const transcript = data?.channel?.alternatives[0].transcript;
  const confidence = data?.channel?.alternatives[0].confidence;
  if (confidence < 0.5 || !data.speech_final) return;
  // my own logic to handle transcripts...
  handleDeepgramTranscript(transcript, participant);
});

[deepgram-community[bot]]

Thanks for asking your question. Please be sure to reply with as much detail as possible so the community can assist you efficiently.
_{Consider joining our Discord community for more opportunity to engage with your fellow Deepgram users. You can earn points which can be redeemed for cool stuff by being active in our communities!}

[deepgram-community[bot]]

Hey there! It looks like you haven't connected your GitHub account to your Deepgram account. You can do this at https://community.deepgram.com - being verified through this process will allow our team to help you in a much more streamlined fashion.

[deepgram-community[bot]]

It looks like we're missing some important information to help debug your issue. Would you mind providing us with the following details in a reply?

• The programming language you are working in (e.g. JavaScript, Python).
• The deepgram product you are using (e.g Speech to Text, Agent API)
• A request ID that triggered your error or issue.